Avander Software Development

EHS systems promise continual improvement. Test yours. Take your last serious near miss and trace the hazard back. Was it really new, or was it already in a risk assessment, a JSA, a permit note or an LMRA? If it was, the information existed, but it never reached the next decision. In development, we treat a known problem that recurs after being fixed as a process failure, and so should EHS too. Three questions every investigation should ask: • Was this hazard or control weakness already known? • Where did the control break down: design, availability, use or checks? • What must change before similar work starts next time? The people closest to the work, contractors included, often know where the process broke. Ask them: "What did you know or notice that didn't make it into the next plan?" We are building tools so each finding is routed back to the relevant risk assessment, JSA, permit or field check, for the responsible people to review and act on. If a risk was written down once, nobody should have to learn it twice.

AI can write a JSA easily, but our EHS partners asked us for something else first. Stop making crews repeat themselves. The same hazards, same controls, same equipment entered again and again because one form doesn't know what the last one already said. Generating documents looks impressive in a demo, but the real question is whether it saves time for the crew on a Tuesday morning. Start with: • the hazards and controls already known for this task • what crews ran into on similar jobs • what has changed and needs checking today • a short, clear view that can be read easily at the point of work That matters especially for contractors navigating your forms and your site for the first time. Their attention should go to the job in front of them, not to retyping information your system already holds. When we got to real implementations, all of our partners asked us for less unnecessary typing before they asked for more AI.

Correct at the desk. Wrong at the workplace. A JSA can be perfect in the system and still miss things the field team encounters on the job. At the point of work, they may find: • different equipment, or equipment in worse condition • blocked access or a changed route • another crew working close by • a contractor crew with different experience and ways of working When the gap is too big, the process must make the safe response easy: Stop. Review. Replan. The JSA may have been correct when it was written. By the time the work starts, it could be out of date. Work as planned vs. work as done. Our job is to make capturing that difference easy, not to score anyone. Record the difference, and start your next JSA closer to reality.

A software tool can move the form, but it can not move the point of work where the LMRA happens, minutes before the work starts. That is its whole value. It is the final check, and the only one to see today's conditions: • weather and access • housekeeping and the condition of the equipment • the job running next to yours • what changed since the permit was issued Our job is to make that check worth stopping for, and to run it on the device the crew already carries. It should surface the known hazards, the controls, and what previous teams met on similar jobs. It should require minimal typing, so the team can focus on the question that matters: • What is new here, right now? A quick check that shows something useful earns attention. One that repeats what was already entered gets treated as paperwork. Ask your teams what would make the LMRA worth stopping for every time. Their answer is our specification.

Ask an IT team what a permit archive is for, and you hear: storage, search, and audit. The experienced EHS leaders we work with call it a record of how the work really gets done. Look at one year of permits for the same type of job, and you find: • the same hazard added by hand • isolations that never reached the standard method • extra controls added on the day • deviations that became normal because nobody asked why The same file says a lot about your contractors and your site. Which crew met which conditions, and what they had to change when they arrived. Bring that history to the next permit, and to the team doing the job. Before the work starts, ask them what they ran into last time, and what is different today. The people doing the job usually know, and it is rarely written down anywhere.

We could digitalise your Permit to Work and LMRA process easily and give you fancy statistics that come with it. Permit volume, LMRA completion, Overdue actions. Working with our partners, we realized that knowing whether the work is really safe takes more. A completed LMRA is not automatically a useful LMRA. A rising permit count can also be a rising paperwork count. Overdue actions sometimes get closed to keep the report clean. The better questions to ask are: • what new hazards were found at the workplace • what conditions had changed since the permit was issued • which jobs were stopped or modified • what came back into the standard risk assessment, the JSA or the permit The last one matters most. When nothing returns to the process, team learnings stay in the field. Where contractors do the work, that gap is even wider. Their observations are the ones most likely to stay outside your system. Would your numbers survive these questions?